In-line Raman spectroscopy and chemometrics for monitoring cocrystallisation using hot melt extrusion

The application of in-line Raman spectroscopy to monitor the formation of a 1:1 cocrystal of ibuprofen (IBU) as a BCS class II drug and nicotinamide as coformer using hot-melt extrusion (HME) was investigated. The process was monitored over different experimental conditions inserting the Raman probe before the extruder die. Partial least square (PLS) was applied as a robust chemometric technique to build predictive models at different levels of chemometric by dividing the experimental data set into calibration and validation subsets. Powder X-Ray diffraction (PXRD) spectra of a set of standard samples were used as calibration to calculate the cocrystal yield from HME experiments regressed by the PLS models. Examination of the full spectra with standard normal variate (SNV) scatter correction with first derivative provided the best fitting goodness and reliability for prediction. Differential scanning calorimetry (DSC) was used as a complementary technique to confirm the composition of the extrudates. Tracking the cocrystal formation throughout the barrel by inserting two Raman probes simultaneously in two different heating zones revealed highly valuable information for understanding the mechanism of cocrystal formation during the HME process.